PERBANDINGAN KNN DAN SVM UNTUK KLASIFIKASI KUALITAS UDARA DI JAKARTA

Abstract

The growth and economic development of a city is one of the factors causing air pollution because air quality has been mixed with various components of chemical compounds such as motor vehicle exhaust gases and factory smoke waste. Data mining is a method to find out information about air pollution in the city of Jakarta. The data mining method used is classification because this method can process air pollution standard index (AQI) parameter data into information that can show the level of air quality per day using the K-Nearest Neighbor algorithm and Support Vector Machine. 
 The result of the application of data mining for air quality classification in Jakarta is that the Support Vector Machine algorithm has better accuracy performance compared to the K-Nearest Neighbor algorithm. The Support Vector Machine algorithm uses the RBF kernel and 100 kernel parameter gets an accuracy value of 98%, precission of 97%, recall of 97%, and F1-Score of 97% while the K-Nearest Neighbor algorithm uses the number of K as much as 6 gets an accuracy value of 96%, precission of 96%, recall of 93%, and F1-Score of 94%.